Big Site SEO Triage (BlueGlassX)

Big Site SEO Triage

Dr. Peter J. MeyersMarketing ScientistSEOmoz

Are you a real doctor?

Why I’m a believer…

My philosophy…

KeywordDensity*

*just kidding

My philosophy…

Q: Is there a dupe content penalty?

My philosophy…

A: I don’t give a shit.

I. Look at the damned site

Your top 2 SEO tools

Scope out the architecture

Navigate the main links

Quick-count internal links

Check for pagination

Scan search filters, sorts…

II. Drill down quickly

Master the basics

Expert’s handiwork

Amateur’s handlesswork

Two useful combos

site: + intitle:site: + inurl:

Quick duplicate check

site:example.comintitle:”unique phrase”

Spot-check URL parameters

site:example.cominurl:page=

Spot-check the architecture

site:example.com/menu

Real-world example

11M - site:example.com8.4M - site:example.com/shop6.7M - site:…/shop inurl:option

How reliable is site:?

(a) site:seomoz.org(b) site:…/blog(c) site:…/blog intitle:spam

Experiment results…

(a) 67,700 – 114,000(b) 4,580 – 8,650(c) 40 – 40

Don’t forget the basics

þ Robots.txtþ HTTP headersþ HTML headers

III. Bring in the firepower

GWT – HTML Improvements

GWT – Index Status (Advanced)

I. Main pagesII. CategoriesIII. Products

GWT – Tiered XML sitemaps

Screaming Frog – URL parameters

SF – Track bad internal links

IV. Choose your weapon

+ Don’t need dev access+ Good for prevention - Still can be dangerous - Bad for removal

Robots.txt

+ Extremely flexible+ Fairly easy to reverse - Occasionally ignored

Meta noindex

+ Still very powerful+ Impact users + search - Difficult to reverse - Google can get suspicious

301 redirects

+ Fast and powerful+ Very flexible - Easy to screw up

Rel=canonical

+ Low-risk option+ Allows pages to rank - Difficult to implement

Rel=prev, Rel=next

Rule #1 for page-level cues

They can’t see whatThey don’t crawl

+ Requires no site access - Inconsistently applied - Google-specific

GWT parameter handling

Technical SEO isn’t dead

Thank You!

Dr. Peter J. [email protected]@dr_pete